JPH0575223B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a halftone image forming method for recording halftone images.

[Prior Art] In image recording devices such as laser printers and injet printers, which can express the density of a recorded image using only two gradations, black and white, or very few gradations, halftone images are Conventionally, halftone images have been recorded using a halftone image generation method called a dither method or a density pattern method.

In the density pattern method, as shown in Figure 6a, one pixel G of the image read from the original image is further divided into n×m
For each micropixel g, the pixel G
By assigning a threshold value for binarizing the image of , and sequentially comparing each threshold value of this threshold matrix pattern with the image density of pixel G, the minute pixel g at the position where the threshold value is larger is "white", and vice versa. “Black” if
A halftone image MG is formed, and this halftone image MG is
is recorded as a halftone image.

On the other hand, in the dither method, n×m
By assigning a threshold similar to the density method to the pixels G, and sequentially comparing this threshold matrix pattern with the density of the image read from the original image in units of n x m, the pixels with the larger threshold are The halftone image is “white” for G, and “black” for the opposite case.
MG is formed, and this halftone image MG is recorded as a halftone image.

In this case, since the number of threshold values in the threshold value matrix pattern is 6×6, the number of intermediate tones that can be expressed as a whole is 36 gradations.

Therefore, it has been conventional practice to apply this principle to record halftone images in multicolor printing.
The density signal obtained by color separation for each cyan color is
If the image is compared with each threshold value of the threshold value matrix pattern as shown in the figure, binarized, and simply overprinted, moiré will appear. Therefore, in order to alleviate such morea, the threshold values of the n×m threshold value matrix patterns are structured to have different screen angles Θ for each separated color, for example, as shown in FIGS. Θ=0 for
By forming a threshold matrix pattern with a screen angle of Θ = 14 degrees for magenta and Θ = 76 degrees for cyan, and by comparing with each threshold matrix pattern, a method for recording halftones in multicolor printing is known. It is being In this case, as a method of forming a halftone image at a desired screen angle, as shown in Japanese Patent Publication No. 52-49361, a seventh A threshold pattern as shown in Figures a to c is selected, and a large matrix MX having one fundamental period L as shown in Figure 8 is extracted by utilizing the periodicity of this threshold pattern in the main scanning direction. It is known that a signal generated by repeating threshold values in a large matrix is compared with the density signal of the original image to form a multicolor halftone image at a desired screen angle.

[Problems to be Solved by the Invention] However, in the method of forming a halftone image at a desired screen angle using the periodicity of the threshold pattern as described above and recording this as a halftone image, the threshold value Since it is necessary to store the pattern for one period of its basic period, there is a problem that the memory capacity becomes large. Also, if you increase the size of the threshold matrix pattern,
Although the number of halftones that can be expressed increases, there is a problem in that the resolution decreases.

The present invention has been made to solve these problems, and provides a halftone image forming method that simultaneously satisfies resolution and gradation and can reduce the memory capacity for storing threshold patterns. The purpose is to

[Means for Solving the Problems] The present invention is configured by having at least two threshold values arranged diagonally with respect to the center as a core, and threshold values whose values increase sequentially around the core. In a halftone image forming method, a halftone image is formed by preparing a threshold matrix pattern having a desired screen angle and comparing an image signal obtained by scanning an original image with the threshold matrix pattern pixel by pixel,
A rectangular unit threshold pattern smaller than the threshold matrix pattern determined from the periodicity of the threshold is stored in a storage means, and the unit threshold patterns stored in the storage are sequentially shifted by a predetermined amount and read out. It is characterized by performing comparison with a threshold value.

[Function] In the present invention, a unit threshold pattern smaller than a threshold matrix pattern determined from the periodicity of the threshold values of the threshold matrix pattern is stored in a storage means, and the unit threshold patterns stored in the storage means are sequentially stored in a predetermined amount. By shifting and reading out, a halftone image is formed.

Thereby, it is possible to provide a halftone image forming method that reduces the memory capacity for storing the threshold matrix pattern and satisfies resolution and gradation at the same time.

[Example] Figure 1a shows an example of the threshold matrix pattern used in the present invention, in which an 8×8 threshold matrix MX is divided into four sub-matrices SMX.
1 to SMX4, and 8 x 8 thresholds are distributed in order from the smallest to each submatrix, and the first to fourth submatrices SMX1 to SMX4 are divided into SMX4.
In SMX4, black pixels grow as shown in FIG. 1b, using threshold values "1", "2", "3", and "4" as nuclei, respectively. Note that this first
In the threshold matrix pattern in the figure, the screen pitch x in the main scanning direction is x=8, and the screen pitch y in the sub-scanning direction is y=2, so the screen angle Θ is based on the rational tangent, Θ=tan ^-1 y/x = tan ^-1 2/8 = 14 degrees.

In this case, by connecting 8×8 thresholds in the main scanning direction, a threshold matrix pattern as shown in FIG. 2 can be developed. As is clear from the figure, the same threshold appears every L=34 in the main scanning direction, and in the sub-scanning direction "36, 59, 61, 49...
42'', ``27, 19, 45, 29...51'', two different patterns of P=2, appear repeatedly. Therefore, L
Either move the threshold pattern consisting of ×P forward by S=8 every time P lines advance in the sub-scanning direction, or
A threshold matrix pattern with a screen angle Θ of 14 degrees can be very easily formed by moving it backward by =26.

Note that L is the length of the threshold pattern, P is the type of threshold pattern in the sub-scanning direction, and D (=L-S).
represents the distance from the starting halftone dot position of the previous scan line, and can be determined as follows.

That is, if the number of dots included in m×n threshold matrix patterns A=x ² ×y ² , then the type of threshold pattern P=GCD(x, y) (however,
GCD (x, y) is the greatest common divisor of x, y), L=
A/P, D=1/V (p・y+t・A) (however, t
is the smallest positive integer such that D is an integer).

In this way, by dividing the threshold matrix into multiple submatrices and arranging the thresholds so that the black pixels in the halftone image spread sequentially from the center of each submatrix, it is possible to avoid reducing the number of gradations. The resolution can be increased to

Further, since the threshold value matrix can be used by circulating L×P threshold values, the memory capacity can be small.

By the way, when the threshold value matrix shown in FIG. 1 is used, when only one halftone dot is recorded, the pitch becomes coarse and the resolution deteriorates. Furthermore, when an odd number of halftone dots are recorded, the distribution of black pixels becomes uneven.

Figures 3 a to d show examples of threshold matrices for solving such problems. Θ
= 45 degree matrix, each of which is a diagonal submatrix SMX1, SMX2, and SMX3 among the four submatrices SMX1 to SMX4.
and the core threshold of SMX4 are set to the same value.

In this way, since at least two black pixels are recorded, the distribution of black pixels is prevented from becoming non-uniform, and it is possible to prevent the pitch from becoming coarse.

In this case, the number of gradations is half of the density method using 8×8 thresholds, but the resolution is twice as high. Also, when compared with the density method using 6 x 6 thresholds, the number of gradations when using the threshold value matrix in Figure 3 is 32 (=16 x 2), and that of the density method is 36
(=6×6), so they are almost the same. but,
The resolution will be doubled. In addition, the number of threshold values is 36, the same as in the density method, and overall, the resolution is improved with less memory capacity.

FIG. 4 is a block diagram showing an embodiment of a circuit for forming a halftone image using the method described above, and the image memory 1 stores image data for y lines each consisting of x and y pixels. x address counter 2 based on the readout clock fx in the main scanning direction.
The stored image data is read out by the x-direction address signal generated by the y-direction address signal generated by the y-address counter 3 based on the readout clock fy in the sub-scanning direction, and is compared via the control gate circuit 4. It is input to circuit 5.

On the other hand, the threshold memory 6 stores one of the threshold patterns for a desired screen angle among the threshold patterns shown in FIGS. 5 a to d when the threshold values in FIGS. 3 a to d are developed in the main scanning direction. has been done. Then, it follows the readout clocks in the main scanning direction and the subscanning direction that are generated by the x address counter 7 and the Y address counter 8 based on the readout clocks fx', fy' having the same frequency as the readout clocks fx, fy of the input image data. The signals are sequentially read out and input to the comparator circuit 7 via the control gate circuit 4.

In this case, the x address counter 7 is preset with a value of "S" for each P line in order to move the LxP thresholds forward by S for each P line. As a result, P types of threshold patterns with length L are sequentially generated and compared with the input image data in the comparator circuit 5. If the threshold value exceeds the image data, the pixels are "white", and if the opposite is the case, "black" halftone dots are generated. Image data is formed.

This halftone image data is recorded by the recording device 9 as a halftone image.

In this case, the threshold memory 6 only needs to have a storage capacity of 36 addresses because the maximum number of thresholds for the screen angle of Θ=45 degrees is 36 among the four types of screen angles. .

Note that in FIGS. 1 and 3, the matrix consisting of 8×8 threshold values is divided into four submatrices, but it may be divided into two.

[Effects of the Invention] As explained above, in the present invention, a unit threshold pattern smaller than a threshold matrix pattern determined from the periodicity of the threshold values of the threshold matrix pattern is stored in the storage means, and Since the halftone image is formed by reading unit threshold patterns sequentially shifted by a predetermined amount, the memory capacity for storing the threshold matrix pattern can be significantly reduced while satisfying resolution and gradation at the same time. It has the effect of being able to

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an example of a threshold matrix pattern used in the present invention, FIG. 2 is a diagram developed in the main scanning direction of the threshold matrix pattern in FIG. 1, and FIG. 3 is a diagram showing a threshold matrix pattern used in the present invention. FIG. 4 is a circuit diagram showing an example of a circuit for forming a halftone image using the method of the present invention, and FIG. FIG. 6 is a diagram showing the threshold matrix pattern of the conventional concentration method and dither method. FIG. 7 is a diagram showing the conventional threshold matrix pattern with screen angle. FIG. 3 is a diagram showing a large matrix of threshold values obtained by repeatedly generating a threshold value matrix pattern of FIG. 1... Image memory, 6... Threshold memory, 5... Comparison circuit.

Claims (1)

[Scope of Claims] 1. Having a desired screen angle configured by having at least two threshold values arranged diagonally with respect to the center as a core, and threshold values having successively larger values arranged around the core. A halftone image forming method in which a threshold matrix pattern is prepared and an image signal obtained by scanning an original image is compared pixel by pixel with the threshold matrix pattern to form a halftone image, comprising: A rectangular unit threshold pattern smaller than the threshold matrix pattern is stored in a storage means, and the image signal and the threshold are compared by sequentially reading out the unit threshold patterns stored in the storage means with a predetermined amount of shift. A halftone image forming method. 2. Claim 1, characterized in that the core thresholds of the threshold matrix pattern are set to the same value, and the threshold matrix pattern is composed of four sub-threshold matrix patterns that are the same in the diagonal direction. halftone image forming method.